Skip to content

Supply the qualifying-individual count in Kentucky dependent care tests - #10976

Merged
MaxGhenis merged 1 commit into
mainfrom
fix-ky-cdcc-test-qualifying-count
Nov 15, 2026
Merged

MaxGhenis merged 1 commit into
mainfrom
fix-ky-cdcc-test-qualifying-count

Conversation

@MaxGhenis

@MaxGhenis MaxGhenis commented Nov 15, 2026 •

Copy link
Copy Markdown
Contributor

Summary

Four cases in ky_cdcc.yaml fail on main. Each expects a federal cdcc above zero and gets 0, as in this job on another PR's merge with main.

#10854 made cdcc_qualified_care_expenses zero unless count_cdcc_eligible exceeds zero. #10324 added Kentucky cases that set capped_count_cdcc_eligible and define no people, so count_cdcc_eligible computes to zero in them.

This PR sets count_cdcc_eligible to the capped count in the five cases that check the federal credit. Four of them fail today. The fifth, the 2021 high-income case, expects a federal credit of 0 and gets it today because the qualified expenses compute to zero, before ARPA's phase-out applies.

The PR changes no formula or parameter.

Tests

Pull request run 43733176655 at 3d311caa4f passed all 36 jobs on GitHub's merge with main e7505bd0ec.

The states-shard-3 baseline job runs ky_cdcc.yaml.

🤖 Generated with Claude Code

The federal credit's qualified expenses are zero unless count_cdcc_eligible
exceeds zero. Five Kentucky cases that check the federal credit set only
capped_count_cdcc_eligible, so four of them fail on main and the fifth
passes for the wrong reason. Set count_cdcc_eligible in each.

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
@MaxGhenis
MaxGhenis merged commit 8dfe48b into main Nov 15, 2026
36 checks passed
@MaxGhenis

Copy link
Copy Markdown
Contributor Author

US + core hub merge audit, us#10976 at 3d311caa4f4b5eb8f24a84d9b4dd5b201d2f2ba9 (squash): Supply the qualifying-individual count in Kentucky dependent care tests.

  • Reviews:
    • The PR has no GitHub reviews. It adds one input to five cases in ky_cdcc.yaml and a changelog fragment, and it changes no formula or parameter.
    • Review pending after merge. Every review lane had closed when I queued the independent review, Subfleet job 20261115-110231-hub-rev-us10976. I will post its result here. If it finds a wrong result, I will revert this change or merge a fix within 24 hours of its return.
    • I merged ahead of the review because four cases in ky_cdcc.yaml fail on main without this change, and they fail the states-shard-3 job of every PR.
  • Gates checked live at merge:
    • gh pr checks exits 0;
    • the latest Pull request run at the head, 43733176655, passed and tested this head merged into main at e7505bd0ec;
    • MERGEABLE;
    • not a draft;
    • no reviewer's latest review requests changes;
    • the semantic-overlap guard: since e7505bd0ec, main has changed no area this PR touches;
    • the merge pins the head.
  • Impact: none. The PR changes test inputs only.
  • Next: 14 other 2021 cases in ky_cdcc.yaml set capped_count_cdcc_eligible without count_cdcc_eligible. They check ky_cdcc_potential alone, and Kentucky's 2021 formula reads the capped count, so the missing count changes none of them.

🤖 Generated with Claude Code

@MaxGhenis
MaxGhenis deleted the fix-ky-cdcc-test-qualifying-count branch November 15, 2026 17:21
MaxGhenis added a commit that referenced this pull request Nov 15, 2026
Since the last merge, main added #10002, #10924 and #10976. None of
them changes this branch's four files.

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
MaxGhenis added a commit that referenced this pull request Nov 15, 2026
Main now carries the Kentucky dependent care credit test fix (#10976).
No commit on main since the last merge changes a file this branch
changes, so the merge has no conflict.

Main adds 2017 to 2020 values to three Massachusetts senior circuit
breaker files and a 2017 value to Michigan's homestead credit cap. All
four files index on gov.irs.uprating, and the model projects 2025 and
later years from each file's latest stored value, so no projected
amount moves.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@MaxGhenis

Copy link
Copy Markdown
Contributor Author

Independent review after merge: approved at 3d311caa4f, with no wrong result found.

  • The reviewer recomputed the five changed cases by hand. Each matches its expected value once the test supplies count_cdcc_eligible.
  • A sweep of the repository's dependent care tests found no other case that passes only because the count computes to zero.
  • 26 U.S.C. 21(a)(1) and 21(b)(2)(A) allow the credit and count expenses only when the unit has a qualifying individual. That supports the gate Correct historical federal taxes and add annual 2017–2026 tests #10854 added.

One follow-up: cases 13 to 26 of ky_cdcc.yaml set capped_count_cdcc_eligible without count_cdcc_eligible. They pass correctly today, because Kentucky's 2021 formula reads the capped count and its own expense inputs. Supplying the count in those cases too would keep them right if that formula ever reads the federal count. I will open a test-only pull request for that.

MaxGhenis added a commit that referenced this pull request Nov 15, 2026
…ests (#10994)

Supply the uncapped count in the other 2021 Kentucky dependent care tests.

Cases 13 to 26 of ky_cdcc.yaml set capped_count_cdcc_eligible without count_cdcc_eligible. They pass because Kentucky's 2021 formula reads the capped count and its own expense inputs, but their inputs described a unit with a capped count of one or two and a count of zero. Set count_cdcc_eligible to the capped count in each, as #10976 did for the five cases that check the federal credit. No expected value changes.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant